Re: [PATCH 3/3] mm/rmap: fix PTE table overruns during batched reference checks
From: Baolin Wang
Date: Thu Nov 12 2026 - 04:39:25 EST
On 11/12/26 4:29 PM, David Hildenbrand (Arm) wrote:
On 11/12/26 08:25, Baolin Wang wrote:
On 11/12/26 6:00 AM, David Hildenbrand (Arm) wrote:
With a folded PMD, pmd_addr_end() simply returns the supplied end.
folio_referenced_one() can therefore walk past the PTE table when
checking references to a large folio.
Folded PMDs are uncommon, but can occur even on arm64, for example
with 64K pages and a 42-bit virtual address space.
Use pte_table_clamped_end() to clamp the range to the PTE table and
VMA boundaries.
Fixes: 52e054f71840 ("mm: rmap: support batched checks of the references for
large folios")
Cc: stable@xxxxxxxxxxxxxxx
Reported-by: Chris Mason <mason@xxxxxxxxxx>
Signed-off-by: David Hildenbrand (Arm) <david@xxxxxxxxxx>
---
LGTM.
Reviewed-by: Baolin Wang <baolin.wang@xxxxxxxxxxxxxxxxx>
BTW, the 'end' calculation for the batch operation in lru_gen_look_around()
could also use the pte_table_clamped_end() helper. But this can be a follow-up.
end = min(addr | ~PMD_MASK, vma->vm_end - 1) + 1;
Cool, can you send a cleanup patch? I'd pick it up once these patches are upstream.
I'll move these into core.git today, so you can send against that for-next later
today.
Sure.